Summary Sperm must swim through mucus before ascending to the upper tract and fertilizing the egg. In addition, sperm must be hyperactivated to penetrate the zona pellucida of the oocyte. Polymeric antibodies (Ab) such as IgM that bind sperm surface antigens can agglutinate sperm into clusters too large to penetrate through the pores of mucus. In addition, IgG can also be engineered to crosslink individual sperm to mucins, as well as inhibit hyperactivation, consequently preventing sperm from reaching and fertilizing the egg. Topical passive immunization based on vaginal delivery of anti-sperm Ab (ASA) was validated in animal models in the 1980s-1990s, and directly overcomes the variable intensity and uncertain reversibility of contraceptive vaccines. However, this strategy was not practical due to the high costs of mAb production and challenges with IgM formulation. Given the remarkable advances in bioprocessing that have greatly reduced the costs of mAb manufacturing, and the possibility of sustained delivery of antibodies in the vagina via intravaginal rings, we believe the time is now ripe to develop novel, ultra-potent ASA for non-hormonal contraception based on topical passive immunization of the vagina against human sperm. In pilot studies, we have shown that we can greatly increase the sperm-agglutination potency of ASA by >15-50-fold by engineering multivalent, IgG-based construct comprised of six to ten Fab domains (i.e. 4 to 8 additional Fabs linked to the parent IgG molecule). These novel constructs possess comparable stability to IgG in accelerated thermal stability studies, and also similar production and purification yield as IgG. One of the constructs represent the lead molecule under development in the Boston University Contraceptive Research Center, with two Phase I clinical trials slated to begin in 2020. In this project, we seek to further improve the potencies by employing cutting edge bispecific Ab engineering technology and affinity maturation via yeast and mamallian cell display to engineer a library of ultra-potent bispecific multivalent ASA constructs. We will target both CD52g, a well characterized and validated antigen target present on human sperm only, and EPPIN, a well established contraceptive target that functions by inhibiting sperm hyperactivation. In Aim 1, we will utilize yeast display to improve the affinity of Fab against CD52g and EPPIN, engineer a library of multivalent bispecific constructs that bind both targets, and rigorously characterize these molecules. In Aim 2, we will perform a panel of studies to assess their potential contraceptive efficacy, including sperm agglutination, trapping individual sperm in mucus, and inhibition of sperm hyperactivation. To determine which construct will most effectively reduce progressive sperm motility in human vagina, we will perform in vivo dose finding studies in Aim 3 with the two most potent constructs from Aims 1 and 2 in a post-coital fertility test adapted to the sheep vagina. Successful completion of these studies will likely allow us to advance a novel ASA candidate with substantially greater contraceptive potency than the current ASA molecules under development for non-hormonal contraception.